Phylotranscriptomic analysis of the origin and early diversification of land plants

Abstract

Reconstructing the origin and evolution of land plants and their algal relatives is a fundamental problem in plant phylogenetics, and is essential for understanding how critical adaptations arose, including the embryo, vascular tissue, seeds, and flowers. Despite advances in molecular systematics, some hypotheses of relationships remain weakly resolved. Inferring deep phylogenies with bouts of rapid diversification can be problematic; however, genome-scale data should significantly increase the number of informative characters for analyses. Recent phylogenomic reconstructions focused on the major divergences of plants have resulted in promising but inconsistent results. One limitation is sparse taxon sampling, likely resulting from the difficulty and cost of data generation. To address this limitation, transcriptome data for 92 streptophyte taxa were generated and analyzed along with 11 published plant genome sequences. Phylogenetic reconstructions were conducted using up to 852 nuclear genes and 1,701,170 aligned sites. Sixty-nine analyses were performed to test the robustness of phylogenetic inferences to permutations of the data matrix or to phylogenetic method, including supermatrix, supertree, and coalescent-based approaches, maximum-likelihood and Bayesian methods, partitioned and unpartitioned analyses, and amino acid versus DNA alignments. Among other results, we find robust support for a sister-group relationship between land plants and one group of streptophyte green algae, the Zygnematophyceae. Strong and robust support for a clade comprising liverworts and mosses is inconsistent with a widely accepted view of early land plant evolution, and suggests that phylogenetic hypotheses used to understand the evolution of fundamental plant traits should be reevaluated.

Keywords: Streptophyta; land plants; phylogenomics; phylogeny; transcriptome.

Fig. 1.

Hypothesized land plant relationships evaluated for all matrices included in this study. Support for these hypotheses was evaluated across all 69 analyses performed in this study. For each hypothesis shown here, a “bar” (e.g., “GA” in the “Sister to land plants” block) indicates unspecified resolution within a grade. Dashed lines indicate lineages or grades with placements that are not relevant to the specified hypothesis. Note that differences with the nonfocal parts of the topology may exist in the given references (see below), and that additional studies may have recovered similar topologies. Abbreviations used: ANA, Amborellales, Nymphaeales, Austrobaileyales grade; Bryo, bryophytes; Char, Charales; Chl, Chloranthales; Col, Coleochaetales; Con, conifers; Cy, cycads; Eudi, eudicots; GA, green algae; Gko, Ginkgo; Gne, Gnetales; Gym, gymnosperms; Hw, hornworts; LP, land plants; Lv, liverworts; Mag, magnoliids; Mo, mosses; Mono, monocots; Pin, Pinaceae; VP, vascular plants; Zygn, Zygnematophyceae. Sister to land plants: Timme et al. (27), Karol et al. (34), Finet et al. (35), Bryophytes: Qiu et al. (42), Chang and Graham (38), Renzaglia et al. (44), Nishiyama et al. (41), Nickrent et al. (40); Gymnosperms: Bowe et al. (92), Lee et al. (52), Chaw et al. (91); Angiosperms: Qiu et al. (12), Burleigh et al. (15), Soltis et al. (14), Soltis et al. (105).

Fig. 2.

ML phylogram inferred from concatenated alignments of first and second codon positions for 674 genes after gene alignments missing more than 50% of taxa were removed and sites gapped in more than 50% of taxa were filtered. Bootstrap values for the concatenated supermatrix analysis are shown on branches (* for 100%) along with the percentages of gene trees exhibiting significant conflict (bootstrap support >75%) with nodes discussed in the text. See Figs. S6 and S7 for bootstrap values and percentages of conflicting gene trees for all nodes in analyses of first and second codon positions and amino acids, respectively. Most nodes exhibited 100% bootstrap support in the concatenated analysis, but gray branches highlight nodes with less than 100% support. Gene models from sequenced genomes are indicated by blue text.

Fig. 3.

Coalescent-based tree estimated from 424 gene trees estimated from first and second codon position alignments after removing genes with less than 50% taxon occupancy after gene fragments missing more than 66% of their sites were removed. Bootstrap values for the ASTRAL analysis are shown on branches (* for 100%) along with the percentages of gene trees exhibiting significant conflict (bootstrap support >75%) with nodes discussed in the text. See Figs. S8 and S9 for bootstrap values and percentages of conflicting gene trees for all nodes in analyses of first and second codon positions and amino acids, respectively. Most nodes exhibited 100% bootstrap support in the ASTRAL analysis, but gray branches highlight nodes with less than 100% support. Gene models from sequenced genomes are indicated by blue text.

Fig. 4.

Summary of support for hypotheses of land plant relationships across 52 supermatrix and coalescent-based analyses including permutations of the full data matrix (Table S2). Occupancy-based gene trimming was carried out by removing genes for which >50% the full taxon set were not included in the alignment. Site trimming removed columns in the aligmment for which >50% of the full taxon set were represented by gap characters. Long-branch trimming was performed on gene trees when a terminal branch was 25-times longer than the median branch length. More stringent, blast-based removal of sequences identified as possible contaminants resulted in a set of 604 gene families (see SI Materials and Methods for stringent filtering strategy). Supermatrix analyses were done with and without partitioning of genes into model parameter classes. Filtering and analysis strategies are indicated below each column and include combinations of: (i) untrim: untrimmed/unfiltered data; (ii) unpart: no data partitions; (iii) 50genes: occupancy-based gene trimming at 50%; (iv) 50sites: occupancy-based site trimming at 50%; (v) gamma: full Gamma (vs. PSR approximation of Gamma); (vi) Chara: gene trimming to exclude genes not present in Chara vulgaris; (vii) 25X: long branch trimming; (viii) 33taxa: sequences with more than 66% gaps in gene alignments removed; (ix) 604genes.trimEXT: aggressive BLAST-based and long-branch filtering of sequence assemblies for each taxon followed by GBLOCKS filtering of sites (SI Materials and Methods). Strong support refers to bootstrap values above 75% for a clade containing the specified taxa. All trees and alignments are available in iPlant’s data store (mirrors.iplantcollaborative.org/onekp_pilot).